Corporate Culture of Beijing Longxin Zhongke Technology Service Center Co., Ltd.

1. Take the road of information preservation.

People often ask me how large foreign companies can do better than those that invest billions of dollars and thousands of people in R&D teams every year, while we only have one hundred people and tens of millions of funds. This problem has also puzzled us for a long time. Later, we found that we were all blinded by some superficial phenomena, so we came to the conclusion that the weak could not defeat the strong. In fact, history develops in the process of the weak constantly defeating the strong. The success or failure of a thing or an organization lies not in whether it is strong at present, but in whether it conforms to the trend of historical development and whether it is supported by the broad masses of the people. Because of this, the weak * * * production party can defeat the powerful Kuomintang and gain political power; Because of this, the weak Intel company was able to beat the powerful IBM company and become the leader in the CPU market in the 1980s. Chairman Mao told us long ago that a single spark can start a prairie fire.

In fact, weakness and strength are relative, and now they are strong. If they lose their advanced nature, they may not be strong in the future. Now the weak, as long as advanced, will gradually become stronger. Future state is determined by current state and future input, which is the theoretical basis of computer. For Loongson, the current status has been determined, and the key is how to input it in the future. In my opinion, as long as the "future input" is correct, Godson's "future state" has the following advantages.

First, adhere to the direction of information serving the broad masses of the people. The development purpose of Loongson processor is: insist on independent innovation, master the core technology in the information field, and serve the national strategic needs through independent, reliable and safe processor design; Insist on informatization to serve the broad masses of people, take the road of economical informatization, and greatly reduce the cost of informatization through the design of high-performance, low-cost and low-power processor.

The nature of foreign monopoly groups' pursuit of huge profits determines that it is impossible for them to realize the informationization of China people. They constantly upgrade their computer products to force users to update the information mode of their computers, which does not meet the needs of people in China and the vast number of third world countries. It is impossible for China to complete the informationization of the people of the whole country in the case of huge profits from western monopoly companies. It is also impossible for the people of China to achieve informatization through the piracy of the whole nation.

At present, more than 50% people in the United States have computers, while there are only 50-60 million computers in China with a population of 654.38+03 billion (these 50-60 million computers are facing the fate of being completely eliminated in order to run Vista). In China, only the informationization of the whole people, including farmers, is the real informationization. Just as China people can't learn from Americans to regard villas as housing standards, we should also seriously consider whether to continue to follow the information road designated by the United States. For example, according to the current model, 500 million computers need to consume the electricity of five Three Gorges power stations. China people can only take the road of informationization with China characteristics and the sustainable development of China economy, and must master the core technologies themselves, so as to greatly reduce the cost of informationization.

The technical route of mainstream PC processors is to continuously use Moore's Law to improve performance within the price range of 150-500 dollars; The technical route of consumer electronics is to continuously use Moore's Law to reduce the cost under the applicable performance. Godson's technical route is to use Moore's Law to continuously improve performance in the price range of 15-25 USD and the power consumption range of 3-5 watts. The application of computers is mainly reflected in two aspects. First, improve production efficiency, mainly including low-end Internet access, office, rural informatization, industrial control and industrial application terminals; And high-end large-scale scientific computing to solve huge challenges. The second is to improve living standards, mainly entertainment including 3D games. According to the national conditions of China, the application orientation of Loongson mainly lies in improving production efficiency. Although Loongson takes low power consumption and low cost as its important design goal and does not pursue the ultimate performance, as a computer CPU, it still needs appropriate performance. For example, most people think that the performance is between 1GHz Pentium III and 2GHz Pentium IV (the performance of Godson 2 is roughly in this range), which is enough for most office Internet applications. At present, in this performance range, Godson has obvious advantages in cost and power consumption.

The advanced nature of Godson is the main driving force to ensure the long-term development of Godson. Due to the above positioning of Loongson, Loongson can adopt the technical route of low cost and low power consumption on the premise of ensuring applicable performance. In contrast, foreign monopoly groups, which aim at profiteering in essence, can't take the road of cost informationization sincerely because they have historical burdens.

Second, the transformation of technology and market is conducive to Loongson's advantage as a latecomer and its leap-forward development. At present, the development of global high-performance general-purpose CPU is facing a technological transition period. The sustained and rapid development of Moore's law; The popularity of the Internet is changing the application mode of computers. The main frequency computer processor design technology is coming to an end; Performance-to-power ratio is becoming the main design index of computer after cost performance. The above technological changes provide an opportunity for China to give full play to its advantages as a latecomer in the next few years, and to find new ways and achieve breakthroughs through leap-forward innovation.

During the Tenth Five-Year Plan period, a leap-forward innovation was realized through the development of Loongson 2, and during the Eleventh Five-Year Plan period, a greater leap-forward innovation will be realized through the development of Loongson 3. If Godson 1 starts in the fog, we can't see the shadow of the forerunner. After the completion of Godson 2, we have surpassed some peers and seen the leader's back. Then the completion of Godson 3 will make us catch up with and surpass the front runners in great strides. Godson 3 will continue to adhere to the technical route of low cost and low power consumption. On the basis of ensuring above GHz and moderately increasing the main frequency, the main design indicators are to reduce power consumption and cost. The power consumption of Godson 3 with 4 cores is 5- 10 watts, and that of Godson 3 with 16 cores is 20-30 watts.

In contrast, foreign monopoly enterprises have a heavy historical burden, and it is difficult to turn around when the ship is big. These historical burdens determine that they can't really embark on the low-cost information road. For example, Intel's latest classmate computer is very chaotic, and it feels that Intel is going to cut costs. As long as you calculate an account, you can see that this is just a spoiler for Intel, provided that the computer does not affect the traditional mainstream computers. Intel sells more than 65438+ billion CPU chips every year, with sales of $35-40 billion. The average price of each chip exceeds $200, and the gross profit exceeds 150. Suppose Intel introduces a low-cost computer CPU at a price of about $20, and the gross profit of each chip is about $5. Annual sales of 200 million chips, sales of $4 billion, gross profit of $654.38+0 billion. If the sales of these 200 million low-cost chips do not affect the sales of the original 654.38 billion chips, Intel is willing to do it; However, if the sales of these 200 million low-priced chips reduce the sales of other processors by 20 million, Intel's sales will be reduced by 4 billion dollars and its profits will be reduced by 3 billion dollars. Therefore, Intel wisely positioned Classmate as an emerging market, and they will never allow this business to affect its mainstream business. The problem is that Intel can really decide the above rules of the game without our ability; But when we have the ability, the rules of the game decided by others' wishful thinking may not be effective. Since we can have a CPU with suitable performance at a price of $20, why should we follow others to play games that are constantly being eliminated and upgraded?

For a large enterprise like Intel, a team of tens of thousands of engineers is both an advantage and a disadvantage. On the one hand, there must be enough profits to support such a team and force Intel to take the high-profit route in the past; On the other hand, once the technical habits and organizational methods of engineers are formed, they will form an inertia. Intel has other historical burdens. For example, Intel tried to get rid of the historical burden of X86 compatibility through the Itanium processor, but it failed, and AMD took advantage of it and was greatly hurt.

Third, although China's industrial environment is controlled by foreign monopoly groups, it is actually beneficial to the development of Loongson. In 2003, the sales revenue of China information industry was 65.438+88 billion yuan, the profit was 75.03 billion yuan, and the profit rate was 3.99%. In 2005, the sales revenue of information industry in China was 384 1 1 billion yuan, the profit was130.7 billion yuan, and the profit rate was 3.4%. It can be said that China's information industry is large but not strong, and it sells high-tech products, but it is actually not a high-tech industry.

On the surface, the current industrial environment is favorable to foreign monopoly groups and unfavorable to Loongson. The backbone enterprises in the domestic information industry (such as Lenovo) are subject to CPU suppliers, and even rely mainly on rebates from CPU suppliers to make profits. They dare not offend CPU providers to use Loongson. It can be said that China's information industry is completely controlled by foreign monopoly groups. But in essence, the above environment is also conducive to the development of Loongson. Thousands of enterprises in China's information industry are selling boxes instead of selling them because they don't master the core technology. Everything is the same when they open the box. The whole industry is transparent and everyone is looking for changes. In particular, a large number of small and medium-sized enterprises struggling to survive in the cracks of large enterprises are very willing to accept such new things as Loongson and seek market opportunities through reform. Moreover, due to the transparency of the industry and fierce competition, as long as a few small and medium-sized enterprises use Loongson to gain a competitive advantage in the market, a large number of other enterprises will flock. As long as you aim at a few breakthroughs, it may have a chain effect. At present, there are such enterprises in the Yangtze River Delta and the Pearl River Delta. They invested boldly, hoping to guide market changes and gain the first chance by using Loongson.

In addition, the domestic software has a solid foundation. After years of development, domestic software has reached the usable stage and has a certain market scale, which needs to be changed from usable to easy to use through large-scale application. At present, it is the best time to combine domestic software and hardware.

Fourth, the Loongson Club is strongly supported by the government, and there is a foundation for sustainable development by relying on computing. The research and development of Loongson has always been supported by the state. So far, the state's investment in Loongson R&D has exceeded 1 billion RMB; The central and local governments have other investments in the application and industrialization of Loongson. As long as we make a simple comparative analysis, we can find that only China government will sincerely support the development of general CPU like Loongson. Last year, a friend told me that he had met the chief designer of E2K processor on the plane (E2K was designed by Russians, but most of the designs were completed around 2000, but there were no slides. At that time, it was the focus of international conferences. Before we designed Loongson 1, Director Li Guojie considered introducing E2K technology to develop general CPU technology in China, but the price was too high. The chief designer told my friend. Now, E2K's project has stopped, and the design team has been acquired by big American companies and then sold.

Backed by the calculation of Chinese Academy of Sciences, Godson is in an invincible position and has the foundation of sustainable development. On the one hand, as a national team, it can continuously provide talents and core technology reserves for Loongson; On the other hand, it provides a rear area for the development of Loongson industrialization, especially in the early stage of industrialization, relying on calculation to carry out Loongson industrialization can make Loongson invincible. For example, because the national plan needs a process to start, the Loongson Research Group has not been supported by the National R&D Fund since 2005. If Loongson had been supported by venture capital, the Loongson team would have been disbanded long ago, but we can get the support of Chinese Academy of Sciences and Computing Institute to make our R&D and industrialization advance normally.

Of course, there are also problems within the system, and we should pay attention to overcoming these problems. The government can only support the research and development of Loongson and the industrialization of "helping the horse to get a ride", and the final success of Loongson must be judged by market behavior.

In short, although the western monopoly groups seem to be strong, they are actually very fragile; Although Godson is still weak, it has strong vitality and bright future. If there is one place in the world that can break the monopoly of foreign monopoly groups on CPU, it must be China. If China has a power to defeat powerful foreign monopoly groups, it must be the power of the people.

2. Take the road of industrialization in which rural areas surround cities.

The combination of universal law and special law is the magic weapon for the success of China Revolution and Loongson industrialization. This has become more and more obvious in the industrialization practice in the past few years.

China government attaches great importance to independent innovation. China's integrated circuit technology and market are developing rapidly, and computer technology and market are transforming. This period is a great opportunity for a team like Loongson to come from behind. This is the universality that Loongson faces, but Loongson also has its particularity.

One of the particularities of Loongson processor is its foundation and universality. High-performance general-purpose CPU chip is the most basic and core chip in the information field, and it is the "Everest" in the chip. At present, only the United States has fully mastered the development technology of high-performance processors. The research and development cycle of high-performance general-purpose CPU is long, so it is difficult for an enterprise to bear the research and development process and expenses independently, and it needs the drive of national will. Different from other special chips (such as MP3 chips), Godson is not only oriented to specific applications, but can be developed and popularized from chip to system in one unit. Godson's versatility leads to its wide application, which can be used in all walks of life, such as computers, network equipment, industrial control, high-definition multimedia and so on. It is impossible for a unit to build an industrial chain one by one, but it must form an industrial environment and industrial cluster of a certain scale. Imagine how many domestic enterprises will know how big the industrial environment of general-purpose processors is if Intel crosses over. Godson has a long industrial chain, which involves not only the development of CPU itself, but also many links such as chipset, motherboard, operating system, application software, system integration and service. Because of this, Godson is not an ordinary chip. The successful industrialization of Loongson will drive the whole industrial cluster and greatly promote the national economy of China. It will directly impact large foreign enterprises like Intel, Microsoft and IBM, and break the information monopoly and digital hegemony of the United States.

The above foundation and universality of Loongson require us to fully understand the arduousness of Loongson industrialization; We are required to build not only the industrial chain, but also the industrial environment and industrial clusters. We are required to win the state's support to improve the industrial environment and build industrial clusters.

Another particularity of Loongson processor is that the market threshold is high, and the market is controlled by powerful international monopoly enterprises. Domestic system integration enterprises are subject to CPU suppliers such as Intel or AMD, and the bigger enterprises (such as Lenovo) are more afraid of offending Intel and AMD. This makes Godson fall into the dilemma of chicken or egg first: on the one hand, it needs Godson to have strong competitiveness and high maturity to have market opportunities; On the other hand, Loongson can constantly understand the market demand and improve its competitiveness and maturity only in the market. Scientific research institutes often take "indomitable spirit" as the requirement of scientific research, which means that the research done has a place in the world; "Site" means that the research done can be accepted by the market, and the relationship between them is "or". The "indomitable spirit" in the field of general CPU has its particularity, and there is a relationship between the two. Only with indomitable spirit can we gain a firm foothold. The rule of the game in the chip market is "winner takes all". As long as it is worse than others 1%, it will lose 100%. Only by reaching the international leading level can we gain a foothold in the market.

Many people have an understanding that things like Loongson cannot be industrialized without the support of the government. Godson's first order must come from government procurement or from large domestic independent innovation enterprises. Our past experience tells us that we need government support, but we can't rely on it. The government certainly supports Loongson, but it can't handle the industrialization of Loongson well. The government can only support the research and development of Loongson and the technical transformation of "helping the horse with a journey". The success of Loongson industrialization, in the final analysis, is to make Loongson itself competitive with its international counterparts. At present, more than a dozen enterprises at home and abroad have adopted Loongson to develop their own products, all of which are purely commercial activities.

Recently, with the opportunity of cooperation between the Chinese and French governments and the cooperation between Italy and France, I have learned a lot from Italian law, especially the correct attitude towards chip products; We have also established an industrialization base in Jiangsu, and with the support of the local government, we have completed the pilot project of 10,000 units. All these have greatly improved the competitiveness and maturity of Loongson. It can be said that after years of hard work, our latest product, Godson 2E, especially Godson 2F, has reached the top level and can compete with international counterparts and gain a foothold.

The particularity of Loongson tells us that we should fully understand the difficulties of Loongson R&D and industrialization, and be full of confidence. We should also be prepared for long-term hard struggle, avoid impetuousness and start from small things in a down-to-earth manner. "Surrounding the city from the countryside" is the correct strategy of Loongson industrialization at present. The so-called "countryside surrounding cities" has two meanings. One is to start from the countryside, that is, from the weak links of opponents; Second, we have a city in mind, that is, our ultimate goal is to realize an independent and controllable information industry.